Computational models of word combinations can help correct inaccuracies in brain activity decoding.

“Why that was important was you can use all those possible sentences to create a computational model, computer model of all the different word combinations to give different sentences given those 50 words, and then you can essentially do what we call autocorrect. It's the same kind of thing that we do when you're texting, for example. You get the wrong letter in there, but your phone actually knows, you know, because of its context what to correct it. So because the decoding's not 100% correct all the time, in fact, it's far from that, it's really helpful to have these other features like autocorrect, the stuff that we use routinely now with texting that makes it correct and then updates it.”